Re: [aur-requests] [PRQ#18963] Deletion Request for dxvk-winelib

On 4/22/20 8:20 AM, Christopher Snowhill (kode54) wrote:

...

On Apr 21, 2020, at 3:28 AM, Stelios Tsampas via aur-requests wrote:

 On 4/21/20 11:53 AM, notify@aur.archlinux.org wrote:

...

soredake [1] filed a deletion request for dxvk-winelib [2]:

https://github.com/doitsujin/dxvk/commit/436357e28096f5e1e25aa8b72fceb77123e...

[1] https://aur.archlinux.org/account/soredake/ [2] https://aur.archlinux.org/pkgbase/dxvk-winelib/

The version pointed by the tag still builds correctly under wine. The commit you are pointing to

is after the tag built by this package. And even so, since this package builds correctly for this version,

what is the problem with it existing? Why should this be deleted?

Are you suggesting that this package maintain a fork of the library which still supports winelib builds? Or instead that it continue to exist to build and install this old commit which will quickly become obsolete as further improvements are added past this commit?

I am suggesting neither. If you bother to look at the commit history, since the tagged version  `1.6.1` there have been two commits at the time of writing, one of which is removing winelib support. There has been no release neither any groundbreaking change. It is rather premature to request to delete a package that build the latest version, especially without contacting the maintainer first about it. Also, allowing it to exist to build the last tagged version with winelib support is not unheard of and it has been done with numerous python packages in the past of example as python2 support was being dropped.